play Namespace Reference

Functions
dict	load_policy (Path filename)

Variables
	policy_path = Path('/home/alex/qi3/cuberl/build/examples/rl/rl_example_10/policy.csv')
dict	policy = load_policy(policy_path)
int	max_episode_steps = 200
str	version = 'v0'
str	env_tag = f"CliffWalking-{version}"
	env
	state = observation
	_
bool	done = False
int	total_reward = 0
dict	action = policy[state]
	observation
	reward
	truncated
	info
bool	is_slippery = False

Function Documentation

load_policy()

dict play.load_policy

(

Path

filename

)

Variable Documentation

_

play._

protected

action

dict play.action = policy[state]

done

bool play.done = False

env

play.env

Initial value:

=  gym.make(id=env_tag,
                   max_episode_steps=max_episode_steps,
                   render_mode="human")

env_tag

str play.env_tag = f"CliffWalking-{version}"

info

play.info

is_slippery

bool play.is_slippery = False

max_episode_steps

int play.max_episode_steps = 200

observation

play.observation

policy

dict play.policy = load_policy(policy_path)

policy_path

play.policy_path = Path('/home/alex/qi3/cuberl/build/examples/rl/rl_example_10/policy.csv')

reward

play.reward

state

play.state = observation

total_reward

int play.total_reward = 0

truncated

play.truncated

version

str play.version = 'v0'